On-Line Bayesian Tree-Structured Transformation of Hidden Markov Models for Speaker Adaptation

This paper presents a new recursive Bayesian learning approach for transformation parameter estimation in speaker adaptation. Our goal is to incrementally transform (or adapt) the entire set of HMM parameters for a new speaker or new acoustic enviroment from a small amount of adaptation data. By establishing a clustering tree of HMM Gaussian mixture components, the finest affine transformation parameters for individual HMM Gaussian mixture components can be dynamically searched. The on-line Bayesian learning technique proposed in our recent work is used for recursive maximum a posteriori estimation of affine transformation parameters. Speaker adaptation experiments using a 26-letter English alphabet vocabulary are conducted, and the viability of the on-line learning framework is confirmed